Database Reference
In-Depth Information
Table 14.1
Number of Detected Visitors at Each of the Venues in the Shopping Mall
During the One-Month Tracking Period, Ranked from High to Low. Venue Names
have been Anonymized According to the Type of Products/Services They Offer
(M: male, F: female)
Venue
Detected Visitors
Venue
Detected Visitors
clothes MF
8064
clothes F 5
378
supermarket
2694
clothes F 2
376
household 3
1964
clothes M 1
354
household 1
1526
snacks sweet
260
clothes knitting
1461
lingerie 2
247
books etc 1
1171
bistro 1
231
clothes F 4
972
clothes F 1
226
mobilephones etc 1
889
bistro 2
199
cosmetics 1
810
clothes M 2
160
shoes
799
interim office
121
hobby
776
optician
101
snacks
717
mobilephones etc 2
93
clothes F 3
704
jewelry
92
home entertainment
673
flowers
75
household 2
667
hair salon
52
lingerie 1
588
leatherware
51
cosmetics 2
575
photo services
41
books etc 2
511
a photo services store that attracted the smallest share of visitors. As is the case
in most shopping malls, one can see that there are a number of dominant anchor
stores accompanied by smaller stores.
As an example on how these tracking data can be mined for interesting
knowledge or patterns, we will focus on
association rules
between the different
shops customers visit in the same shopping trip. As such, the sequence in
which shops were visited is of no importance in this analysis. Additionally,
note that we cannot distinguish between customers who made a purchase in a
store and customers who did not. More formally, the problem can be defined
as follows. Let
I
=
i
1
,i
2
,...i
n
be a set of binary attributes called
item
s. In this
specific case, these items represent a customer's presence in each store. Each
customer's visiting pattern constitutes a
transaction
, which contains a subset
of the items in
I
. An association rule can then be defined as
X
⇒
Y
where
X, Y
⊆
I
and
X
∩
Y
=∅
. The itemsets
X
and
Y
are called
antecedent
and
consequent
respectively. Different measures can be used to select interesting
rules from the set of all possible rules. The
support
of an itemset is defined as
the proportion of transactions in the data set that contain the itemset, and the
support of an association rule is defined as the support of its antecedent. The
